This set of Matlab (7.0) functions contain the core code to reproduce some results of the following two papers:

1. Bayesian Compressive Sensing, Shihao Ji, Ya Xue, and Lawrence Carin, IEEE Trans. Signal Processing, vol. 56, no. 6, June 2008.
2. Multi-Task Compressive Sensing, Shihao Ji, David Dunson, and Lawrence Carin, IEEE Trans. Signal Processing, vol. 57, no. 1, pp. 92-106, Jan. 2009.

• Fig2.m ---> generate Fig.2

  • The following two Matlab files from l1-magic are required for BP implementation:
  • l1qc_logbarrier.m
  • l1qc_newton.m
  • which can be found at: http://www.acm.caltech.edu/l1magic/

• Fig4_ab.m ---> generate Figs.4(a,b)

  • multi_random_measures.m ----> generate the "Random" curve for Fig.4a
  • multi_optimized_measures.m ----> generate the "Optimized" curve for Fig.4a
  • multi_approx_measures.m ----> generate the "Approx." curve for Fig.4a

• Fig2.m ---> generate Fig.2

• Fig3.m ---> generate Fig.3

  • multi_runs_75.m ----> generate the "MT 75%" curve for Fig.3
  • multi_runs_50.m ----> generate the "MT 50%" curve for Fig.3
  • multi_runs_25.m ----> generate the "MT 25%" curve for Fig.3

A bug was fixed on Aug. 03, 2008 in MT_CS.m for the cases where signals are dramatic undersampled.

Distribution and use of this code is subject to the following agreement:

This Program is provided by Duke University and the authors as a service to the research community. It is provided without cost or restrictions, except for the User's acknowledgement that the Program is provided on an "As Is" basis and User understands that Duke University and the authors make no express or implied warranty of any kind. Duke University and the authors specifically disclaim any implied warranty or merchantability or fitness for a particular purpose, and make no representations or warranties that the Program will not infringe the intellectual property rights of others. The User agrees to indemnify and hold harmless Duke University and the authors from and against any and all liability arising out of User's use of the Program.

For bug reports, please contact Shihao Ji at [email protected].

Shihao Ji Duke University, July 8, 2007